Thirty-one patients with knee effusions associated with rheumatoid arthritis (RA) have been treated with two intraarticular (i.a.) injections of each 330 mg sodium morrhuate (SM) used for synoviorthesis versus a single injection of 20 mg triamcinolone hexacetonide (TA). During an observation period of one year, five articular parameters as well as patient's and doctor's global assessments were evaluated. TA showed an earlier onset and a longer duration of therapeutic effects with high statistical significance. The maximum improvement was significantly more pronounced with TA than with SM. Finally after one year improvement measured by a remission index was observed in 81% versus 33% resp. of all joints injected. Due to ineffectiveness of the primary treatment nine patients (60%) out of the SM group, but not patient out of the TA group had to be crossed over to the other treatment. SM usually caused a reactive effusion within hours after injection requiring arthrocentesis. In conclusion efficacy and tolerability are clearly better for TA than for SM.